#38 erf function eccentric?

[Anonymous]

I thought I'd use the erf() function for a quick calc
concerning outliers to a standard distrib. but didn't
get the result I expected for my quick rule of thumb
(~95% falls within +-2 STD) check. Looking further it
seems you use a different function definition than,
say, Shaum Mathematical hand book and, as a result,
you have to take the square-root of the argument if it
is expressed in STDs. I think this might be an error.

Shaum integrates exp(-t^2/2) cf Easy calc exp(-t^2)